You keep saying "defensive ships will cost more but be more effective defenders." But the question, really, is "if two civs spend the same amount, one on offensive and the other on defensive, which civ will generally be in a better position?" If the offensive civ will be, then we've got no change from the GC2 dynamic. If the defensive civ will be, we have turtling.
Why does it have to be all of one or all of the other? Let's say you want to go to war with someone. You can go full-on offense, and assume that no enemy offensive ships will get past your blockades. If your assumptions are wrong, you're going to lose worlds.
If defense is sufficiently strong, everyone is going to want some minimal level of protection. That's why your "remove the orbit cap" doesn't work; it requires you to invest too much into static defense around a single world to make defense meaningful.
Minimal defenses are your protection against a fast-strike. They keep your enemy honest; in order to take your planet, they need to be serious about it. They can't just throw a unit or two at it and get it for free; they're going to have to work for it.
Those who choose not to have this minimal protection gain a slight economic advantage, at the potential cost of losing worlds. There should be varying degrees of defense, each degree requiring more and more economy to build and maintain. Yes, there should be a breaking point where the defense isn't worth it. But in GC2, that breaking point is one ship; it's never worthwhile to have a ship on (static) defense.
The best news, is that it is OPTIONAL! So those who don't care to use the additional tactical options, or prefer not to mess with all the military mumbo-jumbo, don't have to.
No, that's actually terrible news. People think that being able to switch something on and off is good for a game. It isn't; options always have a cost.
Balance requires knowing what all the factors are. You can't effectively make a balancing decisions without knowing all of the minutae that goes into that balance. If those details change, if the user can decide what those details are, it is not very hard to get into situations where things work when the option is set one way, and not when it is set another way.
For example, there is an entire Super Ability devoted to making colonizing of certain hazardous worlds easier. This ability is meaningless if you turn such worlds off. A simple option screws up the balance. But it can screw things up in more significant ways too.
What if the idea behind having hazardous worlds was so that the overall pacing of the game would be broken up by a mini-colony rush, one that could be initiated by whomever felt the need to expand. That is, it's an alternative means of expanding after the colony rush without having to go to war. It allows a wily player to come back from behind by keeping the big boys squablling, while he goes off and takes some choice territory.
If that strategy were important to the overall balance of the game, being able to turn it off has an affect on game balance.
Note that this example is about something that was added to the game in an expansion. Combat is absolutely fundamental to the game; putting an imbalance there is the kiss of death.
A good tactical combat system is about as far removed from "bigger number wins" as it gets. Fleet composition, special effect modules, relative ship sizes, maneuvering; all of these should play a part in a good tactical combat system.
Such a complex system is necessarily difficult-to-impossible to simulate well. If the simulation AI isn't as good at combat as the player could be, then you create a fundamental imbalance. The player who decides to play the battles has an economic advantage. So you can either play through these battles and spend less on military, or spend less time on something that you don't particularly like, but have a harder time playing the game overall.
This gets even worse when combined with multiplayer. Do I dare pit a mere AI against a human? So now, I have to fight each tactical battle, even if I don't want to. That will make multiplayer games drag on even more than they would anyway.
As I mentioned before, a good tactical combat system requires a lot of front-work care and feeding. Proper fleet composition and all of that. That's just what you need to do to get your fleet to the battle. So, it creates this incredible complexity in ship and fleet development. This is very annoying for the person who's simulating tactical combat, because he's got to deal with this complexity that a more simplified combat system, like one used for non-tactical combat TBSs, would not have required.
It's a big can of worms. Oh, you can certainly make a TBS that has optional tactical combat. But will your game be better off overall for doing so than taking one road over another?
The best way to do this is to effectively make two separate games. Make the main game non-tactical, and then release an expansion that changes everything: resourcing models, tech trees, etc. That version would require tactical battles (possibly with a simulation option for fights that truly don't matter). Such a game would focus more on combat, so there would be fewer victory conditions.
Acsension was a cool idea, but they should make all victory conditions be scored more evenly in the end game, so you don't get shafted score-wise for going the non-military route.
There's a fundamental problem with that. A military victory, pretty much by definition, means that you own most of the worlds in the galaxy. The other victory conditions don't require this. So long as score increases with the number of planets you have, that will be the best way to increase your score.